aix+oracle tuning+1
TRANSCRIPT
Advanced Technical Skills
August 29, 2012© 2012 IBM Corporation
AIX Performance: Configuration & Tuning for Oracle
Vijay [email protected] - Oracle Solutions Team
Advanced Technical Skills
2August 29, 2012© 2012 IBM Corporation
Legal information
The information in this presentation is provided by IBM on an "AS IS" basis without any warranty, guarantee or assurance of any kind. IBM also does not provide any warranty, guarantee or assurance that the information in this paper is free from errors or omissions. Information is believed to be accurate as of the date of publication. You should check with the appropriate vendor to obtain current product information.
Any proposed use of claims in this presentation outside of the United States must be reviewed by local IBM country counsel prior to such use.
IBM, ^, , RS6000, System p, AIX, AIX 5L, GPFS, and Enterprise Storage Server (ESS) are trademarks or registered trademarks of the International Business Machines Corporation.
Oracle, Oracle9i and Oracle10g are trademarks or registered trademarks of Oracle Corporation.
All other products or company names are used for identification purposes only, and may be trademarks of their respective owners.
Advanced Technical Skills
3August 29, 2012© 2012 IBM Corporation
� AIX Configuration Best Practices for Oracle
– Memory
– CPU
– I/O
– Network
– Miscellaneous
Agenda
Advanced Technical Skills
4August 29, 2012© 2012 IBM Corporation
� The suggestions presented here are considered to
be basic configuration “starting points” for
general Oracle workloads
� Your workloads may vary
� Ongoing performance monitoring and tuning is
recommended to ensure that the configuration is
optimal for the particular workload characteristics
AIX Configuration Best Practices for Oracle
Advanced Technical Skills
7August 29, 2012© 2012 IBM Corporation
� AIX Configuration Best Practices for Oracle
– Memory
– CPU
– I/O
– Network
– Miscellaneous
Agenda
Advanced Technical Skills
8August 29, 2012© 2012 IBM Corporation
AIX Memory Management Overview
� The role of Virtual Memory Manager (VMM) is to provide the capability for programs to address
more memory locations than are actually available in physical memory.
� On AIX this is accomplished using segments that are partitioned into fixed sizes called “pages”.
– A segment is 256M
– default page size 4K
– POWER 4+ and POWER5 can define large pages, which are 16M
� The 32-bit or 64-bit address translates into a 52-bit or 80-bit virtual address
– 32-bit system : 4-bit segment register that contains a 24-bit segment id, and 28-bit offset.
• 24-bit segment id + 28-bit offset = 52-bit VA
– 64-bit system: 32-bit segment register that contains a 52-bit segment id, and 28-bit offset.
• 52-bit segment id + 28-bit offset = 80-bit VA
� The VMM maintains a list of free frames that can be used to retrieve pages that need to be
brought into memory.
– The VMM replenishes the free list by removing some of the current pages from real memory (i.e., steal memory).
– The process of moving data between memory and disk is called “paging”.
� The VMM uses a Page Replacement Algorithm (implemented in the lrud kernel threads) to select
pages that will be removed from memory.
Advanced Technical Skills
10August 29, 2012© 2012 IBM Corporation
Memory Tuning Overview
Virtual Memory
(General)
Large Pages
(Pinned Memory 1)
Memory:
�minfree
�maxfree
�lru_file_repage
�lru_poll_interval
�v_pinshm
�lgpg_regions
�lgpg_size
JFSEnhanced JFS
(JFS2)
�maxperm
�strict_maxperm
�maxclient
�strict_maxclient
NAME CUR DEF BOOT MIN MAX UNIT TYPE
--------------------------------------------------------------------------------
lru_file_repage 1 1 1 0 1 boolean D
lru_poll_interval 0 0 0 0 60000 milliseconds D
maxclient% 80 80 80 1 100 % memory D
maxfree 1088 1088 1088 8 200K 4KB pages D
maxperm% 80 80 80 1 100 % memory D
minfree 960 960 960 8 200K 4KB pages D
strict_maxclient 1 1 1 0 1 boolean D
strict_maxperm 0 0 0 0 1 boolean D
minperm% 20 20 20 1 100 % memory D
vmo –p –o <parameter name>=<new value>
-p flags updates /etc/tunables/nextboot
Advanced Technical Skills
11August 29, 2012© 2012 IBM Corporation
� The AIX “vmo” command provides for the display and/or
update of several parameters which influence the way AIX
manages physical memory
– The “-a” option displays current parameter settings
� vmo –a
– The “-o” option is used to change parameter values
� vmo –o minfree=1440
– The “-p” option is used to make changes persist across a reboot
� vmo –p –o minfree=1440
Virtual Memory Manager (VMM) Tuning
On AIX 5.3, number of the default “vmo” settings are not optimized for
database workloads and should be modified for Oracle environments
Advanced Technical Skills
12August 29, 2012© 2012 IBM Corporation
Kernel Parameter Tuning – AIX 6.1/7.1
� AIX 6.1/7.1 configured by default to be ‘correct’ for most
workloads.
� Many tunable are classified as ‘Restricted’:
– Only change if AIX Support says so
– Parameters will not be displayed unless the ‘-F’ option is used for commands like vmo, no, ioo, etc.
� When migrating from AIX 5.3 to 6.1/7.1, parameter override
settings in AIX 5.3 will be transferred to AIX 6.1/7.1
environment
Advanced Technical Skills
14August 29, 2012© 2012 IBM Corporation
General Memory Tuning
Memory Use 9/20/2007
0
10
20
30
40
50
60
70
80
90
100
08:0
2
08:0
6
08:1
0
08:1
4
08:1
8
08:2
2
08:2
6
08:3
0
08:3
4
08:3
8
08:4
2
08:4
6
08:5
0
08:5
4
08:5
8
09:0
2
09:0
6
09:1
0
09:1
4
09:1
8
09:2
2
09:2
6
09:3
0
09:3
4
09:3
8
09:4
2
09:4
6
09:5
0
09:5
4
09:5
8
10:0
2
10:0
6
10:1
0
10:1
4
Process% FScache%
� Two primary categories of memory pages: Computational and File System
� AIX will always try to utilize all of the physical memory available (subject to
vmo parameter settings)
– What is not required to support current computational page demand will tend to be used for filesystem cache
– Raw Devices and filesystems mounted (or individual files opened) in DIO/CIO mode do not use filesystem cache
Advanced Technical Skills
15August 29, 2012© 2012 IBM Corporation15
� JFS2 utilizes two caches - one for inodes and one for metadata
� Caches grow in size until maximum size is reached before cache slots are reused
� Default values are tuned for a file server!
� Each entry in the inode cache requires about 1KB of physical memory
� 1MB of memory can cache about 1000 files
� Configured via ioo parameters:
– j2_inodeCacheSize (Default: 400 = 10%) *1
– j2_metadataCacheSize (Default: 400 = 4%) *1
� The current memory use can be verified via:cat /proc/sys/fs/jfs2/memory_usage
metadata cache: 31186944inode cache: 34209792total: 65396736
JFS2 inode / metadata caches
Inode cache
metadata cache
AIX “pinned”
“System memory”
10% *1
4% *1
Process
File cache
Unused
100%
physical memory
Can not be paged
!
*1 Note: Default values in AIX 7.1 are 200 (5%) , 200 (2%)
Advanced Technical Skills
21August 29, 2012© 2012 IBM Corporation
0%
20%
40%
60%
80%
100%
Time
Ph
ys
ica
l M
em
ory
numperm% comp% Free% maxperm%maxfree minfree minperm%
Virtual Memory Management (VMM) Thresholds
Start stealing pages when
free memory below minfree
Stop stealing pages when
free memory above maxfree
When numperm% >
maxperm%, steal only file
system pages
When minperm% <
numperm% < maxperm%,
steal file system or
computation pages,
depending on repage rate
When numperm% <
minperm%, steal both file
system and computational
pages
Advanced Technical Skills
24August 29, 2012© 2012 IBM Corporation
Page Steal Method
� Historically, AIX maintained a single LRU list which contains
both computational and filesystem pages.
– In environments with lots of computational pages that you want to keep in memory, LRUD may have to spend a lot of time scanning the LRU list to find an eligible filesystem page to steal
� AIX 6.1 introduced the ability to maintain separate LRU lists
for computational vs. filesystem pages.
– Also backported to AIX 5.3
� New page_steal_method parameter
– Enabled (1) by default in 6.1/7.1, disabled (0) by default in 5.3
– Requires a reboot to change
– Recommended for Oracle DB environments
Advanced Technical Skills
25August 29, 2012© 2012 IBM Corporation
Large Segment Aliasing (AIX 6.1 TL06, AIX 7.1)
� Feature allows user applications to “automagically” use 1TB
segments.
– 1 SLB entry in POWER7 can now address 1TB of memory.
• Segment Lookaside Buffer (SLB) fault issue no longer relevant
• Immediate performance boost for applications, new and legacy
� Significant changes under the covers
– New address space allocation policy
• Attempts to group address space requests together to facilitate 1TB aliasing.
– Once certain allocation size thresholds have been reached, OS automatically aliases memory with 1TB aliases.
• 256MB segments still exist for handling IO
� Aliasing only available for shared memory regions at this point.
Advanced Technical Skills
27August 29, 2012© 2012 IBM Corporation
Recommended vmo “Starting Points” - Review
90(*3)
0
0
0
1
1
1088
960
10
N/A
0
1
90
90
3
AIX 7.1
Default
No
No
No
No
Yes
Yes
No
No
Yes
N/A
Yes
Yes
Yes
Yes
No
AIX 7.1
Restricted
Parameter Recommend
Value
AIX 6.1
Default
AIX 6.1
Restricted
AIX 5.3
Default
minperm% 3 3 No 20
maxperm% 90 90 Yes 80
maxclient% 90 90 Yes 80
strict_maxclient 1 1 Yes 1
strict_maxperm 0 0 Yes 0
lru_file_repage 0 0 Yes 1 or 0(*1)
lru_poll_interval 10 10 Yes 10
minfree 960 960 No 960
maxfree 1088(*2) 1088 No 1088
page_steal_method 1 1 Yes 0
memory_affinity 1 1 Yes 1
v_pinshm 0 0 No 0
lgpg_regions 0 0 No 0
lgpg_size 0 0 No 0
maxpin% Leave at Default 80(*3) No 80
*1 Depending on AIX 5.3 TL level *2 Do not reduce below default *3 Depends on LSA use – LSA active � 90, otherwise 80
Advanced Technical Skills
28August 29, 2012© 2012 IBM Corporation
� Memory cards are associated with every Multi Chip Module (MCM), Dual Core Module (DCM) or Quad Core Module (QCM) in the server
– The Hypervisor assigns physical CPUs to a dedicated CPU LPAR (or shared processor pool) from one or more MCMs, DCMs or DCMs
– For a given LPAR, there will normally be at least 1 memory pool for each MCM, DCM or QCM that has contributed
processors to that LPAR or shared processor pool
� By default, memory for a process is allocated from memory associated with the processor that caused the page fault.
� Memory pool configuration is influenced by the VMO parameter “memory_affinity”
– Memory_affinity=1 means configure memory pools based on physical hardware configuration (DEFAULT)
– Memory_affinity=0 means configure roughly uniform memory pools from any physical location
� Number can be seen with ‘vmstat –v |grep pools’
� Size can only be seen using KDB
� LRUD operates per memory pool
p590 / p595 MCM Architecture
Understanding Memory Pools
Advanced Technical Skills
30August 29, 2012© 2012 IBM Corporation
AIX Multiple Page Size Support
� 4K (Default)
– All memory pages in the filesystem cache are 4K in size!
– Can be paged to paging space
– 4K pages can be combined to a 64K page if not enough 64K pages are available
� 64K, available with POWER5+ and later & AIX 5.3 TL4+
– Can be paged to paging space
– Can be converted to 4K pages if not enough 4K pages are available
– Kernel page size used in AIX 5.3 TL4+ and above (can be configured)
– Can be utilized for application code, data and stack as well, but requires specific
configuration
� 16M available with POWER4 hardware (or later) (also referred to as Large Pages)
– Requires pinned memory and explicit configuration
– Can not be paged to paging space
� 16G available with POWER5+ & AIX 5.3 TL4+
– Cannot be used with Oracle
– Can not be paged to paging space
Advanced Technical Skills
31August 29, 2012© 2012 IBM Corporation
0
500
1000
1500
2000
2500
3000
3500
4000
4500
5000
14
:02
14
:02
14
:02
14
:03
14
:03
14
:03
14
:03
14
:04
14
:04
14
:04
14
:05
14
:05
14
:05
14
:06
14
:06
14
:06
14
:06
14
:07
14
:07
14
:07
14
:08
14
:08
14
:08
14
:09
14
:09
14
:09
14
:09
14
:10
14
:10
14
:10
14
:11
14
:11
14
:11
14
:12
14
:12
14
:12
14
:12
14
:13
14
:13
14
:13
14
:14
14
:14
14
:14
14
:15
14
:15
14
:15
14
:15
14
:16
14
:16
14
:16
14
:17
14
:17
14
:17
14
:18
Time
MB
us
ed
4KB_used MB 4KB_free MB 64KB_used MB 64KB_free MB 16MB_usedMB 16MB_freeMB
4K - 64K - 16MB Page Dynamics
4KB pages
64KB pages
16MB pages
4kb used
4kb free
64kb used
64kb free
Advanced Technical Skills
32August 29, 2012© 2012 IBM Corporation
AIX Multiple Page Size Support� User/Application must request preferred page size
– 64K page size is very promising, since they do not need to be configured/reserved in advance or pinned• export
LDR_CNTRL=DATAPSIZE=64K@TEXTPSIZE=64K@STACKPSIZE=64K@SHMPSIZE=64K to use the 64K pagesize for stack, data & text
– Will require Oracle to explicitly request the page size (10.2.0.4 & up plus Oracle patch# 7226548)
– If preferred size not available, the largest available smaller size will be used• Current Oracle versions will end up using 64KB pages even if
SGA is not pinned
� Refer: http://www-03.ibm.com/systems/resources/systems_p_os_aix_whitepapers_multiple_page.pdf
Advanced Technical Skills
33August 29, 2012© 2012 IBM Corporation
Large Page Support 16mb (optional)
Pinning shared memory
� AIX Parameters
• vmo –p –o v_pinshm = 1• Leave maxpin% at the default of 80% unless the SGA exceeds 77% of real memory
– Vmo –p –o maxpin%=[(total mem-SGA size)*100/total mem] + 3
� Oracle Parameters
• LOCK_SGA = TRUE
Enabling Large Page Support
� vmo –p –o lgpg_size = 16777216 –o lgpg_regions=(SGA size / 16 MB)
Allowing user oracle to use Large Pages
� chuser capabilities=CAP_NUMA_ATTACH,CAP_BYPASS_RAC_VMM,CAP_PROPAGATE oracle
Using Monitoring Tools
� svmon –G
� svmon –P
Oracle metalink note# 372157.1
Note: It is recommended not to pin SGA, as long as you had configured the VMM, SGA & PGA properly.
Advanced Technical Skills
34August 29, 2012© 2012 IBM Corporation
Determining SGA size
SGA Memory Summary for DB: test01 Instance: test01 Snaps: 1046 -1047
SGA regions Size in Bytes
------------------------------ ----------------
Database Buffers 16,928,210,944
Fixed Size 768,448
Redo Buffers 2,371,584
Variable Size 1,241,513,984
----------------
sum 18,172,864,960
lgpg_regions = 18,172,864,960 / 16,777,216 = 1084 (rounded up)
Advanced Technical Skills
35August 29, 2012© 2012 IBM Corporation
Oracle Memory Structures Allocation
� 11g : Automatic Memory Management (AMM)
– memory_target (dynamic parameter) – specifies the total memory size to be used by the instance SGA and PGA. Exchanges between SGA and PGA are done according to workload requirements
– If sga_target and pga_aggregate_target are not set, the policy is to give 60% of memory_target to the SGA and 40% to the PGA.
– memory_max_target (static parameter) – specifies the maximum memory size for the database instance.
– To use Automatic Memory Management, memory_target must be >0.
– Can not be used together with “LOCK_SGA=TRUE
� See Metalink notes 443746.1 and 452512.1 explaining AMM and these new parameters.
AMM dynamic resizing of the shared pool can cause a fair amount of “cursor: pin s” wait time. One strategy to minimize this is to set minimum sizes for memory areas you particularly care about.
In addition you can change the frequency how often AMM analyzes and adjusts the memory distribution. See: Metalink note: 742599.1 ( _memory_broker_stat_interval)
Advanced Technical Skills
36August 29, 2012© 2012 IBM Corporation
SGA_MAX_SIZE and LOCK_SGA implications (11g, 10.2.4.0+)
� LOCK_SGA=false Preferred
– Oracle dynamically allocates memory for the SGA only as needed up to the size specified by SGA_TARGET
– SGA_TARGET may be dynamically increased, up to SGA_MAX_SIZE
– 64K pages automatically used for SGA if supported in the environment. If needed, 4K (or 16M) pages are converted to 64K pages.
� LOCK_SGA=true Discouraged
– Oracle Automatic Memory Management (AMM) cannot be used (MEMORY_TARGET)
– Oracle pre-allocates all memory as specified by SGA_MAX_SIZE and pins it in memory, even if it’s not all used (i.e. SGA_TARGET < SGA_MAX_SIZE)
– If sufficient 16M pages are available those will be used. Otherwise, all the SGA memory will be allocated from 64K (if supported) or 4K pages (if 64K pages are not supported). If needed, 4K (or 16M pages will be converted to 64K pages, but 16M pages are never automatically created.
– If a value for sga_max_size is specified larger than the amount of available memory for computational pages, the system can become unresponsive due to system paging.
– If the specified SGA_MAX_SIZE is much larger than the currently available pages on the combined 64K and 16M page free lists, the database startup can fail with error: “IBM AIX RISC System/6000 Error: 12: Not enough space”. In this case re-try to start the database.
Advanced Technical Skills
37August 29, 2012© 2012 IBM Corporation
AIX dynamic LPAR with Oracle AMM
Advanced Technical Skills
38August 29, 2012© 2012 IBM Corporation
Memory Usage in an Oracle Environment
Computational� Some used for AIX kernel processing� Some used by Oracle/client executable programs� Includes Oracle SGA and PGA memory
File System Cache � May be used for caching or prefetching of Oracle .dbf files
– Only for local filesystem based (non-RAC) environments where Direct I/O (or Concurrent I/O) is not used
� May be used for other Oracle related files
– Archive logs, export/import files, backups, binaries, etc.
� May be used for non-Oracle related files
– Application files, system files, etc.
Virtual Memory Management Priorities� Always want to keep computational pages in memory --
System paging/swapping may degrade Oracle/application performance
– Allocate enough physical memory to support computational footprint requirement + small file cache
– When necessary, steal filesystem pages, not computational
Advanced Technical Skills
39August 29, 2012© 2012 IBM Corporation
Oracle Server Architecture – Memory Structures
Source: Oracle® Database Concepts
11g Release 1 (11.1)
Part Number B28318-05
Advanced Technical Skills
40August 29, 2012© 2012 IBM Corporation
Displaying Memory Usage StatisticsThe ‘vmstat’ command provides information on current memory usage: ( reported in 4K pages)
# vmstat -v
1048576 memory pages
1002006 lruable pages
812111 free pages
1 memory pools
141103 pinned pages
80.0 maxpin percentage
3.0 minperm percentage
90.0 maxperm percentage
3.2 numperm percentage
32779 file pages
0.0 compressed percentage
0 compressed pages
0.0 numclient percentage
90.0 maxclient percentage
0 client pages
0 remote pageouts scheduled
0 pending disk I/Os blocked with no pbuf
0 paging space I/Os blocked with no psbuf
2484 filesystem I/Os blocked with no fsbuf
0 client filesystem I/Os blocked with no fsbuf
0 external pager filesystem I/Os blocked with no fsbuf
Advanced Technical Skills
41August 29, 2012© 2012 IBM Corporation
Displaying Memory Usage StatisticsThe ‘svmon -G’ command provides information on current memory usage per page size: (general
numbers are reported in 4K pages)
# svmon -G
size inuse free pin virtual
memory 1179648 926225 290287 493246 262007
pg space 1572864 5215
work pers clnt other
pin 91390 0 0 74176
in use 258573 4316 335656
PageSize PoolSize inuse pgsp pin virtual
s 4 KB - 477713 5215 94606 141175
m 64 KB - 7552 0 4435 7552
L 16 MB 80 0 0 80 0
Advanced Technical Skills
42August 29, 2012© 2012 IBM Corporation
AIX Paging Space
Allocate Paging Space:
� Configure Server/LPAR with enough physical memory to satisfy memory requirements
� With AIX demand paging, paging space does not have to be large
� Provides safety net to prevent system crashes when memory overcommitted.
� Generally, keep within internal drive or high performing SAN storage
Monitor paging activity:
� vmstat -s
� sar -r
� nmon
Resolve paging issues:
� Reduce file system cache size (MAXPERM, MAXCLIENT)
� Reduce Oracle SGA or PGA (9i or later) size
� Add physical memory
Do not over commit real memory!
Advanced Technical Skills
43August 29, 2012© 2012 IBM Corporation
A few important notes
AIX 6.1 kernel pinning :
� vmo -r -o vmm_klock_mode=2
� It is enabled by default in 7.1 and can be enabled on 6.1 TL6.
� There were cases of Oracle RAC node eviction due to higher memory allocation and/or bug/feature
USLA heap issue:
� With the release of 11gR2, Oracle introduced a new feature called “on line patching” and is supported on AIX 6.1 TL02 and 7.1 TL0 onwards
� Upgrade to AIX 6.1 TL07 or AIX 7.1 TL01, and then apply the Oracle bug patch 13443029 for database 11.2.0.2 and 11.2.0.3
� If you cannot upgrade to AIX 6.1 TL07 or 7.1 TL01, then apply Oracle patch 10190759, which will have a reduced online-patch functionality
� It is recommended to use AIX 6.1 TL07 SP02 or AIX 7.1 TL01 SP02
P7 special patch:
� Patch 6784747 - Wasted memory in koh-kghu allocations. Free extents of memory might not get used.
� This bug impacts 10.2.0.4/5 and 11.1.0.x releases. This fix is not included in any of the PSU. This issue is fixed in Oracle 11.2.0.1 onwards
Advanced Technical Skills
44August 29, 2012© 2012 IBM Corporation
� AIX Configuration Best Practices for Oracle
– Memory
– CPU
– I/O
– Network
– Miscellaneous
Agenda
Advanced Technical Skills
45August 29, 2012© 2012 IBM Corporation
CPU Considerations
Oracle Parameters based on the # of CPUs
– DB_WRITER_PROCESSES
– Degree of Parallelism
• user level
• table level
• query level
• MAX_PARALLEL_SERVERS or AUTOMATIC_PARALLEL_TUNING (CPU_COUNT * PARALLEL_THREADS_PER_CPU)
– CPU_COUNT
– FAST_START_PARALLEL_ROLLBACK – should be using UNDO instead
– CBO – execution plan may be affected; check explain plan
Advanced Technical Skills
47August 29, 2012© 2012 IBM Corporation
Lparstat command# lparstat -i
• Node Name : erpcc8• Partition Name : -• Partition Number : -• Type : Dedicated• Mode : Capped• Entitled Capacity : 4.00• Partition Group-ID : -• Shared Pool ID : -• Online Virtual CPUs : 4• Maximum Virtual CPUs : 4• Minimum Virtual CPUs : 1• Online Memory : 8192 MB• Maximum Memory : 9216 MB• Minimum Memory : 128 MB• Variable Capacity Weight : -• Minimum Capacity : 1.00• Maximum Capacity : 4.00• Capacity Increment : 1.00• Maximum Physical CPUs in system : 4• Active Physical CPUs in system : 4• Active CPUs in Pool : -• Unallocated Capacity : -• Physical CPU Percentage : 100.00%• Unallocated Weight : -
Advanced Technical Skills
48August 29, 2012© 2012 IBM Corporation
AIX 5.2
LPAR
SMT=off
AIX 5.3
LPAR
SMT=on
AIX 5.3
Micro
Partition
SMT=on
AIX 6.1
Micro
Partition
SMT=off
AIX 7.1Micro
PartitionSMT-4
V V V V V V V V V
13 CPU Shared Processor Pool*
L L L L L L L L L L L L L L
2.1 Proc. Units 0.8 Proc Units 1.2 Proc Units
2 CPUs
(dedicated)
1 CPU
(dedicated)
16 CPU SMP Server
Think “PVL “ P=Physical V=Virtual L=Logical (SMT)
Physical, Logical, Virtual Layers
* All activated, non-dedicated CPUs are automatically placed into the shared processor pool.Only 2.1+0.8+1.2 = 4.1 processor units of “desired capacity” has been allocated from the pool of 13 CPUs
Virtual
L L Logical
PhysicalShared Pool 0 Pool 1
L L L L L L
Advanced Technical Skills
49August 29, 2012© 2012 IBM Corporation
Virtual Shared Processor Pools - Benefits
LPAR1
i5/OS
LPAR2
AIX
Oracle
DB
LPAR3
Linux
Physical Shared Pool (9 processor cores)
1 1 1 1 2 3 4 5 6 7 8 9
LPAR4
Uncapped
AIX
Oracle DB
VP = 5
Ent. = 2.5
LPAR5
Uncapped
AIX
Oracle DB
VP = 4
Ent. = 1.70
LPAR6
Uncapped
AIX
OAS
App 1
VP = 4
Ent. = 2.00
LPAR7
Uncapped
AIX
OAS
App2
VP = 6
Ent. = 2.00
LPAR8
Uncapped
AIX
OAS
App 3
VP = 3
Ent. = 1.00
POWER6/7 Multiple shared pools:
• Can reduce the number of software licenses by putting a limit on the amount of processors an uncapped partition can use
• Up to 64 shared pools
Virtual Shared pool #1
Max Cap: 5 processors
Virtual Shared pool #2
Max Cap: 6 processors
Oracle cores to license:• 1 from dedicated partition LPAR2• 5 from pool 1= 6
Server with 12 processor cores
OAS cores to license:• 6 from pool 2= 6
Oracle DB core – license factors:
Power5 and earlier: 0.75
Power6: 1.0
Power7: 1.0
Advanced Technical Skills
50August 29, 2012© 2012 IBM Corporation
Virtual Processor - Folding� Dynamically adjusting active Virtual Processors (VPs)
– System consolidates loads onto a minimal number of VPs
• Scheduler computes utilization of VPs every second– If VPs needed to host physical utilization is less than the current active VP count, a VP is
put to sleep– If VPs needed are greater than the current active VPs, more are enabled
– On by default in AIX 5.3 ML3 and later
• vpm_xvcpus tunable
• vpm_fold_policy tunable
� Increases processor utilization and affinity
– Inactive VPs don’t get dispatched and waste physical CPU cycles
– Fewer VPs can be more accurately dispatched to physical resources by the Hypervisor with potential for improved processor cache efficiency
� When to adjust – Check with IBM support before changing!
– Burst/Batch workloads with short response-time requirements may need sub-second dispatch latency
• Disable processor folding or manually tune the number of VPs
– # schedo –o vpm_xvcpus=[-1 | N]– Where N specifies the number of VPs to enable in addition to the number of VPs needed to
consume physical CPU utilization– A value of “-1” disables CPU folding
Advanced Technical Skills
53August 29, 2012© 2012 IBM Corporation
CPU: Compatibility Matrix
DLPARSMTMicro-Partition
AIX 5.3
AIX 6.1
AIX 7.1
Oracle 9i
Oracle 10g
Oracle 11g
Note: Live partition mobility is certified for 10.2.0.4 and AIX 5.3 TL8 SP4 & 6.1 TL2 SP3.
Oracle RAC 10.2.0.3 on VIOS 1.3.1.1 & AIX 5.3 TL07 and higher are certified.
Advanced Technical Skills
54August 29, 2012© 2012 IBM Corporation
� AIX Configuration Best Practices for Oracle
– Memory
– CPU
– I/O
– Network
– Miscellaneous
Agenda
Advanced Technical Skills
55August 29, 2012© 2012 IBM Corporation
Application memory area caches data to
avoid IO
NFS caches file attributes
NFS has a cached filesystem for NFS clients
The AIX IO stack
JFS and JFS2 cache use extra system RAM
JFS uses persistent pages for cache
JFS2 uses client pages for cache
Queues exist for both adapters and disks
Adapter device drivers use DMA for IO
Disk subsystems have read and write cache
Disks have memory to store commands/data
IOs can be coalesced (good) or split up (bad) as they go thru the IO stack
Write cache Read cache or memory area used for IO
Disk
Disk subsystem (optional)
Adapter Device Drivers
Disk Device Drivers
LVM (LVM device drivers)
Raw
dis
ks
Raw
LV
s
Application
Logical file system
JFS JFS2 NFS Other
VMM
Multi-path IO driver (optional)
Advanced Technical Skills
58August 29, 2012© 2012 IBM Corporation
AIX Filesystems Mount options� Journaled File System (JFS)
Better for lots of small file creates & deletes
– Buffer caching (default) provides Sequential Read-Ahead, cached writes, etc.
– Direct I/O (DIO) mount/open option � no caching on reads
� Enhanced JFS (JFS2)
Better for large files/filesystems
– Buffer caching (default) provides Sequential Read-Ahead, cached writes, etc.
– Direct I/O (DIO) mount/open option � no caching on reads
– Concurrent I/O (CIO) mount/open option � DIO, with write serialization disabled
• Use for Oracle .dbf, control files and online redo logs only!!!
� GPFS
Clustered filesystem – the IBM filesystem for RAC
– Non-cached, non-blocking I/Os (similar to JFS2 CIO) for all Oracle files
GPFS and JFS2 with CIO offer similar performance as Raw Devices
Advanced Technical Skills
59August 29, 2012© 2012 IBM Corporation
�Direct IO (DIO) – introduced in AIX 4.3.
• Data is transfered directly from the disk to the application buffer, bypassing the file buffer cache hence avoiding double caching (filesystem cache + Oracle SGA). • Emulates a raw-device implementation.
�To mount a filesystem in DIO$ mount –o dio /data
�Concurrent IO (CIO) – introduced with JFS2 in AIX 5.2 ML1
• Implicit use of DIO. • No Inode locking : Multiple threads can perform reads and writes on the same file at the same time. • Performance achieved using CIO is comparable to raw-devices.
�To mount a filesystem in CIO:$ mount –o cio /data
Bench throughput over run duration – higher
tps indicates better performance.
AIX Filesystems Mount options (Cont’d)
Advanced Technical Skills
60August 29, 2012© 2012 IBM Corporation
Cached vs. non-Cached (Direct) I/O
Set filesystemio_options=SETALL
-or-
Use “cio” mount option
Use “cio” mount optionJFS2
Set filesystemio_options=SETALL
-or-
Use “dio” mount option
Set filesystemio_options=SETALL
-or-
Use “dio” mount option
JFS
Oracle 10g/11gOracle 9i
File System caching tends to benefit heavily sequential workloads with low write content. To enable caching for JFS/JFS2:
�Use default filesystem mount options
�Set Oracle filesystemio_options=ASYNC
DIO tends to benefit heavily random access workloads and CIO tends to benefit heavy update workloads. To disable JFS, JFS2 caching, see the
following table:
Note: Refer Metalink Note #s 272520.1, 257338.1, 360287.1, 232935.1
Advanced Technical Skills
61August 29, 2012© 2012 IBM Corporation
CIO/DIO implementation Advices
mount -o rw mount -o rw
mount -o rw mount -o cio *(1)
Cached by Oracle
Cached by AIX
mount -o rw mount -o cio (jfs2 + agblksize=512)
Cached by Oracle
Cached by AIX
mount -o rw mount -o rbrw
Use JFS2 write-behind Y
but are not kept in AIX Cache.
mount -o rw mount -o rw
Oracle bin and shared lib.
with Standard mount options with optimized mount options
Cached by AIX Cached by AIX
Oracle Datafiles
Oracle Redolog
Cached by Oracle
Cached by Oracle
Cached by AIXCached by AIX
Cached by AIXOracle Archivelog
Oracle Control files
*(1) : to avoid demoted IO : jfs2 agblksize = Oracle DB block size / nFlash Recovery Area mount = rbrw
Advanced Technical Skills
62August 29, 2012© 2012 IBM Corporation
CIO Demotion and Filesystem Block Size
Data Base Files (DBF)
� If db_block_size = 2048 � set agblksize=2048
� If db_block_size >= 4096 � set agblksize=4096
Online redolog files & control files
� Set agblksize=512 and use CIO or DIO
Mount Filesystems with “noatime” option
� AIX/Linux records information about when files were created and last modified as well as last accessed. This may lead to significant I/O performance problems on often accessed files such as the contents of the $ORACLE, /tmp.
Advanced Technical Skills
63August 29, 2012© 2012 IBM Corporation
I/O Tuning (ioo)
READ-AHEAD (Only applicable to JFS/JFS2 with caching enabled)
MINPGAHEAD (JFS) or j2_minPageReadAhead (JFS2)
– Default: 2
– Starting value: MAX(2,DB_BLOCK_SIZE / 4096)
MAXPGAHEAD (JFS) or j2_maxPageReadAhead (JFS2)
– Default: 8 (JFS), 128 (JFS2)
– Set equal to (or multiple of) size of largest Oracle I/O request
• DB_BLOCK_SIZE * DB_FILE_MULTI_BLOCK_READ_COUNT
Number of buffer structures per filesystem:
NUMFSBUFS:
– Default: 196, Starting Value: 568
j2_nBufferPerPagerDevice (j2_dynamicBufferPreallocation replaces)
– Default: 512, Starting Value: 2048
Monitor with “vmstat –v”
Advanced Technical Skills
64August 29, 2012© 2012 IBM Corporation
Data Layout for Optimal I/O Performance
Stripe and mirror everything (SAME) approach:
�Goal is to balance I/O activity across all disks, loops, adapters, etc...
�Avoid/Eliminate I/O hotspots
�Manual file-by-file data placement is time consuming, resource intensive and iterative
Use RAID-5 or RAID-10 to create striped LUNs (hdisks)
Create AIX Volume Group(s) (VG) w/ LUNs from multiple
arrays, striping on the front end as well for maximum
distribution
�Physical Partition Spreading (mklv –e x) –or-
�Large Grained LVM striping (>= 1MB stripe size)
http://www-1.ibm.com/support/techdocs/atsmastr.nsf/WebIndex/WP100319
Advanced Technical Skills
70August 29, 2012© 2012 IBM Corporation
Other I/O Stack Tuning Options (Device Level)
lsattr/chdev:
� num_cmd_elems = maximum number of outstanding I/Os for an adapter.
� queue_depth = the maximum # of outstanding I/Os for an hdisk.
Recommended/supported maximum is storage subsystem dependent.
� max_xfer_size = the maximum allowable I/O transfer size (default is 0x40000 or 256k).
Maximum supported value is storage subsystem dependent. Increasing value (to at
least 0x200000) will also increase DMA size from 16 MB to 256 MB.
� dyntrk = When set to yes (recommended), allows for immediate re-routing of I/O
requests to an alternative path when a device ID (N_PORT_ID) change has been
detected.
� fc_err_recov = When set to “fast_fail” (recommended), if the driver receives an RSCN
notification from the switch, the driver will check to see if the device is still on the
fabric and will flush back outstanding I/Os if the device is no longer found.
Advanced Technical Skills
73August 29, 2012© 2012 IBM Corporation
• Allows multiple requests to be sent without to have to wait until the disk subsystem has completed the physical IO.
• Utilization of asynchronous IO is strongly advised whatever the type of file-system and mount option implemented (JFS, JFS2, CIO, DIO).
IO : Asynchronous IO (AIO)
�Posix vs Legacy
Since AIX5L V5.3, two types of AIO are now available : Legacy and Posix. For the moment, the Oracle code is using the Legacy AIO servers.
aioQ
Application aioserversDisk
1
2
3 4
Advanced Technical Skills
74August 29, 2012© 2012 IBM Corporation
IO : Asynchronous IO (AIO) fastpath
• FS with CIO/DIO and AIX 5.3 TL5+ :
� Activate fsfast_path (comparable to fast_path but for FS + CIO/DIO)
AIX 5L : adding the following line in /etc/inittab: aioo:2:once:aioo –o fsfast_path=1
AIX 6.1/7.1 : ioo –p –o aio_fsfastpath=1 (default setting)
• Raw Devices / ASM :
� check AIO configuration with : lsattr –El aio0 enable asynchronous IO fast_path. :
AIX 5L : chdev -a fastpath=enable -l aio0 (default since AIX 5.3)AIX 6.1/7.1 : ioo –p –o aio_fastpath=1 (default setting)
Application Disk
1
2
3
With fast_path, IO are queued directly from the application into the LVM layer without any
“aioservers kproc” operation.
� Better performance compare to non-fast_path� No need to tune the min and max aioservers� No ioservers proc. => “ps –k | grep aio | wc –l” is not relevent, use “iostat –A” instead
AIX Kernel
Advanced Technical Skills
75August 29, 2012© 2012 IBM Corporation
Asynchronous I/O for filesystem environments]
Monitor Oracle usage:
• Watch alert log and *.trc files in BDUMP directory for warning message:
“Warning “lio_listio returned EAGAIN”
� If warning messages found, increase maxreqs and/or maxservers
Monitor from AIX:
• “pstat –a | grep aios”
• Use “-A” option for NMON
• iostat –Aq (new in AIX 5.3)
Advanced Technical Skills
76August 29, 2012© 2012 IBM Corporation
GPFS I/O Related Tunables� Refer Metalink note 302806.1
Async I/O:
� Oracle parameter filesystemio_options is ignored
� Set Oracle parameter disk_asynch_io=TRUE
� Prefetchthreads= exactly what the name says
– Usually set prefetchthreads=64 (the default)
� Worker1threads = GPFS asynch I/O
– Set worker1threads=550-prefetchthreads
� Set aio maxservers=(worker1threads/#cpus) + 10
Other settings:
� GPFS block size is configurable; most will use 512KB-1MB
� Pagepool – GPFS fs buffer cache, not used for RAC but may be for binaries. Default=64M
mmchconfig pagepool=100M
Advanced Technical Skills
77August 29, 2012© 2012 IBM Corporation
I/O Pacing
� I/O Pacing parameters can be used to prevent large I/O
streams from monopolizing CPUs
– System backups (mksysb)
– DB backups (RMAN, Netbackup)
– Software patch updates
� When Oracle ClusterWare is used, use AIX 6.1/7.1 Defaults:
– chgsys -l sys0 -a maxpout=8193 –a minpout=4096 (AIX defaults)
– nfso –o nfs_iopace_pages=1024 (AIX defaults)
– On the Oracle clusterware set : crsctl set css diagwait 13 –force
• This will delay the OPROCD reboot time to 10secs from 0.5secs during node eviction/reboot, just enough to write the log/trace files for future diagnosis. Metalinknote# 559365.1
Advanced Technical Skills
78August 29, 2012© 2012 IBM Corporation
ASM configurations
AIX parameters
– Async I/O needs to be enabled, but default values may be used
ASM instance parameters
– ASM_POWER_LIMIT=1
Makes ASM rebalancing a low-priority operation. May be changed dynamically. It is common to set this value to 0, then increase to a higher value during maintenance windows
– PROCESSES=25+ 15n, where n=# of instances using ASM
DB instance parameters
– disk_asynch_io=TRUE
– filesystemio_options=ASYNCH
– Increase Processes by 16
– Increase Large_Pool by 600k
– Increase Shared_Pool by [(1M per 100GB of usable space) + 2M]
Advanced Technical Skills
79August 29, 2012© 2012 IBM Corporation
� AIX Configuration Best Practices for Oracle
– Memory
– CPU
– I/O
– Network
– Miscellaneous
Agenda
Advanced Technical Skills
80August 29, 2012© 2012 IBM Corporation
Network Options (no) Parameters
– Set sb_max >= 1 MB (1048576)
– Set tcp_sendspace >= 262144
– Set tcp_recvspace >= 262144
– Set rfc1323=1
If isno=1, check to see if settings have been overridden at the network interface level:
$ no -a | grep use_isno=1
use_isno=1
$ lsattr -E -l en0 -H
attribute value description
rfc1323 N/A
tcp_nodelay N/A
tcp_sendspace N/A
tcp_recvspace N/A
tcp_mssdflt N/A
Advanced Technical Skills
81August 29, 2012© 2012 IBM Corporation
Additional Network (no) Parameters for RAC:
� Set udp_sendspace = db_block_size * db_file_multiblock_read_count
(not less than 65536)
� Set udp_recvspace = 10 * udp_sendspace
– Must be < sb_max
� Increase if buffer overflows occur
� Ipqmaxlen=512 for GPFS environments
� Use Jumbo Frames if supported at the switch layer
Examples:
� no -a |grep udp_sendspace
� no –o -p udp_sendspace=65536
� netstat -s |grep "socket buffer overflows"
Advanced Technical Skills
82August 29, 2012© 2012 IBM Corporation
� AIX Configuration Best Practices for Oracle
– Memory
– I/O
– Network
– Miscellaneous
Agenda
Advanced Technical Skills
83August 29, 2012© 2012 IBM Corporation
Miscellaneous parameters
� User Limits (smit chuser)
– Soft FILE size = -1 (Unlimited)
– Soft CPU time = -1 (Unlimited)
– Soft DATA segment = -1 (Unlimited)
– Soft STACK size -1 (Unlimited)
– /etc/security/limits
� Maximum number of PROCESSES allowed per user (smit chgsys)
– maxuproc >= 2048
� Environment variables:
– AIXTHREAD_SCOPE=S
– LDR_CNTRL=DATAPSIZE=64K@TEXTPSIZE=64K@STACKPSIZE=64K
Advanced Technical Skills
89August 29, 2012© 2012 IBM Corporation
Marketplace Myths – Quotes from Customers
� “Oracle DB is only going to run on Sun in the futureY”
� “Oracle will stop supporting Siebel on Power”
� “JD Edwards will stop releasing product on i OS on Power hardware”
� “Oracle DB and Oracle Apps will no longer run on Power hardware”
� “IBM’s Power is no longer relevant to Oracle”
� “POWER systems are losing support across the tech industry”
� “Oracle is dropping their System z strategy and support”
� “Oracle will move away from System z and it should be expected System z will not be
supported in the future”
� “System z is the last platform to get an Oracle Patchset Update”
� “Oracle releases E-Business Suite fixes on Solaris first”
� “Oracle DB is only going to run on Sun in the futureY”
� “Bugs take longer to get fixed on Power. OEL or RHEL are fixed first”
� “Oracle will move away from System z and it should be expected System z will not be
supported in the future”
� “IBM Power is a Tier 4 port for Fusion Middleware”
“I heard thatYY.
Advanced Technical Skills
90August 29, 2012© 2012 IBM Corporation
� Oracle continues to leverage IBM Power technology advantages
– Micropartitioning
• Sub capacity licensing support (exclusive)
– Simultaneous Multi-Threading (SMT)
– Use of VIOS for RAC and non RAC deployments
– Support for Advanced Memory Expansion
– Oracle DB 11gR2 support for Live Partition Mobility (LPM) on AIX 5.3, 6.1, 7.1
• IBM Exclusive (No other hardware vendor, including Sun, supports LPM)
– PowerHA Support
– IBM GPFS
– FMW 11gR1 released concurrently using IBM JDK 6
• Oracle participated in IBM JDK 6 beta project
– InfiniBand
Oracle Certification Of Power Advanced Functionality
Shows partnership
Advanced Technical Skills
91August 29, 2012© 2012 IBM Corporation
AIX Certification Is Better than OEL� Oracle Linux is the laggard
– On Sunday 6 February 2011, Oracle Linux 6 was released
– Oracle Linux is Oracle’s Development Platform for all Oracle Database, Middleware, and Application Products
– As of January 12, 2012 (11 months later), zero Oracle products are certified on Oracle Linux 6
� AIX is the leader
– AIX 7 GA date was in Sept. 2010
– Oracle Certified DB 11gR2 on AIX 7 in Oct. 2010 (30 days later)
– 1 year later, E-Business Suite, PeopleSoft, Oracle Enterprise Manager, Fusion Middleware, Tuxedo, Hyperion EPM, OBIEE, Policy Automation, and others are certified on AIX 7
Oracle sellers will tell customers that it takes a long time for products to become available on AIX, when in fact the exact opposite is the case. Products were delivered
on the latest version of AIX (7.1) more than 11 months sooner than Oracle Linux 6 (6.x). They’re still waiting and waiting and waiting for their first product on OL 6Y
What Really Drives the Oracle – IBM Partnership
Advanced Technical Skills
93August 29, 2012© 2012 IBM Corporation
22%
5% 5%
13%55%
OracleHWPeopleFacilitiesOther
Average Estimated 2011 Oracle SW Solution Costs
Based on averages of customer supplied estimates
Advanced Technical Skills
94August 29, 2012© 2012 IBM Corporation
Advanced Technical Skills
95August 29, 2012© 2012 IBM Corporation
Advantage Of What Or Whom?
Advanced Technical Skills
© 2012 IBM Corporation
Exadata – engineered for Oracle or Customer ?
96
× Market positioning does not lend to control Oracle pricing, service quality and
may lead to lock-in
× No hope of escaping long-term licensing and maintenance fees
× No easy integration into your existing infrastructure strategy
× No support for non-DB Workloads
× No virtualization
× No RedHat or SUSE, Only Oracle’s Linux
× No public benchmarks
× No significant cost reductions as compared to Power alternatives
× No roadmaps (remember HP Itanium did not have good roadmap)
× No granular expansion
× No in-chassis upgrades to future technologies
× No capacity-on-demand or utility pricing
× No third-party best-of-breed integration
× No workload optimization
× No versions of Oracle DB earlier then 11gR2
× No reduction in risk
× No Hardware based RAID, so host based ASM is the only option
All the computers and cars are also
engineered !!
IBM POWER7® Systems Oracle Exadata
1. Open vs. Proprietary
Allows the utilization of “best of breed” solution components from IBM and non IBM solution providers. Build the best solution to meet workload requirements
Customer has little choice in solution design. Forced to buy components not relevant to workloads. One size fits all” requires total trust in Oracle as a single source provider.
2. Real Performance
Power 7 and optimized storage are better at:
1) Data with write requirements (OLTP)
2) DW with real time updates
3) Complex workloads (reads & writes)
4) Concurrent workloads sharing resources
Exadata works best if data is read intensive, pre sorted, bulk loaded or can fit entirely into storage server cache. “Smart Scan”functionality not relevant for indexed tables or OLTP workloads
3. Flexible vs. Rigid
Power 7 has proven value in:
1) Consolidation of Oracle DB instances
2) Resource virtualization across instances
3) Virtualization of DB and non DB workloads
4) Granular resource upgrades and CoD
Exadata has significant limitations in:
1) Consolidation of DB instances
2) Resource virtualization across instances
3) Running N-1 Oracle sw levels
4) Upgrading hw resources granularly
4. StorageIntelligent Storage Management: Flash copy, Remote Mirror, San Volume Controller, RAID, Easy Tier, Concurrent Maintenance, No SPOF
Basic cheap disk – no data management. No internal disk RAID Parity –ASM mirroring only,
No concurrent maintenance, Multiple SPOF
5. Integration Into Existing
Environment
Easily integrated into existing production, test, backup – recovery, storage replication and DR environments
Requires redesign of operational architecture strategy and deployment and new, complex DBA skills in many customer environments
6. ComplexityLess systems to manage
Less OS, DB and RAC and images to update
RAC is an option, not a requirement
Significantly more systems, software images and RAC nodes to manage and update
7. Solution Cost
Initial acquisition costs are similar
Consolidation favours Power (virtualization)
Upgrades cost less on Power
Cost of integration less (existing Power)
Power price / performance costs declining
Initial acquisition costs similar
Contractual terms can favour Oracle
Upgrades cost more on Exadata
Cost of integration more (existing Power)
Price /performance costs increasing
Advanced Technical Skills
98August 29, 2012© 2012 IBM Corporation
Operating System on x86 vs. AIX on Power
� The study correlated the technical components with their associated business metrics that in turn can be used to understand the advantages and key strategies that will help an organization to choose an optimal operating system
� When all of the data is examined, the different OS options were articulated using a graded scorecard approach1
1. Graded on a normal college-level type of curveSource: Does Your OS Matter? Selecting a Strategic Operating System; Solitaire Interglobal Ltd (All rights reserved); October 2011.
AIX on Power Systems has demonstrated strengths and add significant real
business value.
Linux is the OS in Exadata !!
Advanced Technical Skills
99August 29, 2012© 2012 IBM Corporation
Thank
You
Advanced Technical Skills
100August 29, 2012© 2012 IBM Corporation
The following are trademarks of the International Business Machines Corporation in the United States and/or other countries. For a complete list of IBM Trademarks, see www.ibm.com/legal/copytrade.shtml: AS/400, DBE, e-business logo, ESCO, eServer, FICON, IBM, IBM Logo, iSeries, MVS, OS/390, pSeries, RS/6000, S/30, VM/ESA, VSE/ESA, Websphere, xSeries, z/OS, zSeries, z/VM
The following are trademarks or registered trademarks of other companies
Lotus, Notes, and Domino are trademarks or registered trademarks of Lotus Development CorporationJava and all Java-related trademarks and logos are trademarks of Sun Microsystems, Inc., in the United States and other countriesLINUX is a registered trademark of Linux TorvaldsUNIX is a registered trademark of The Open Group in the United States and other countries.Microsoft, Windows and Windows NT are registered trademarks of Microsoft Corporation.SET and Secure Electronic Transaction are trademarks owned by SET Secure Electronic Transaction LLC.Intel is a registered trademark of Intel Corporation* All other products may be trademarks or registered trademarks of their respective companies.
NOTES:
Performance is in Internal Throughput Rate (ITR) ratio based on measurements and projections using standard IBM benchmarks in a controlled environment. The actual throughput that any user will experience will vary depending upon considerations such as the amount of multiprogramming in the user's job stream, the I/O configuration, the storage configuration, and the workload processed. Therefore, no assurance can be given that an individual user will achieve throughput improvements equivalent to the performance ratios stated here.
IBM hardware products are manufactured from new parts, or new and serviceable used parts. Regardless, our warranty terms apply.
All customer examples cited or described in this presentation are presented as illustrations of the manner in which some customers have used IBM products and the results they may have achieved. Actual environmental costs and performance characteristics will vary depending on individual customer configurations and conditions.
This publication was produced in the United States. IBM may not offer the products, services or features discussed in this document in other countries, and the information may be subject to change without notice. Consult your local IBM business contact for information on the product or services available in your area.
All statements regarding IBM's future direction and intent are subject to change or withdrawal without notice, and represent goals and objectives only.
Information about non-IBM products is obtained from the manufacturers of those products or their published announcements. IBM has not tested those products and cannot confirm the performance, compatibility, or any other claims related to non-IBM products. Questions on the capabilities of non-IBM products should be addressed to the suppliers of those products.
Prices subject to change without notice. Contact your IBM representative or Business Partner for the most current pricing in your geography.
References in this document to IBM products or services do not imply that IBM intends to make them available in every country.
Any proposed use of claims in this presentation outside of the United States must be reviewed by local IBM country counsel prior to such use.
The information could include technical inaccuracies or typographical errors. Changes are periodically made to the information herein; these changes will be incorporated in new editions of the publication. IBM may make improvements and/or changes in the product(s) and/or the program(s) described in this publication at any time without notice.
Any references in this information to non-IBM Web sites are provided for convenience only and do not in any manner serve as an endorsement of those Web sites. The materials at those Web sites are not part of the materials for this IBM product and use of those Web sites is at your own risk.
Trademarks